The specific aims of this randomized, double-blinded, placebo-controlled study are to: 1) measure the relative systemic availability of inhaled dexamethasone and assess the consistency of systemic absorption during a course of inhaled dexamethasone; 2) evaluate and compare changes in ventilatory support (rate of ventilation), peak inspiratory pressure, and fractional inspired oxygen in ventilator-dependent premature neonates receiving a 7-day course of inhaled dexamethasone or intravenous dexamethasone and; 3) define relationships between systemic drug exposure and improvements in ventilatory parameters and adverse effects after both intravenous and inhaled dexamethasone. Twenty patients have been enrolled on this study, but results have not yet been analyzed. An important treatment objective is to determine if one or the other forms of therapy being evaluated will reduce the severity and complications of bronchopulmonary dysplasia by minimizing ventilator-induced barotrauma and oxygen toxicity. Previous studies suggesting benefit from intravenous dexamethasone infusions were not blinded, and thus management may not have been equal between treated and untreated groups. Despite the apparent advantage of aerosolized steroids in minimizing systemic steroid exposure, studies conducted in neonates have not compared efficacy, adverse effects, or systemic absorption of inhaled steroids, and compared the outcome to more commonly administered intravenous dexamethasone. If inhaled steroids result in improved pulmonary function comparable to that claimed for intravenous steroids, the risk-benefit ratio would clearly favor treatment by inhalation. This protocol will address this very important issue in the care of premature neonates.